Search

We investigate nonradial $f$-mode oscillations of hybrid neutron stars in full general relativity, employing hybrid equations of state describing a nuclear outer core and a pasta-phase transition to a quark-matter core. The validity of various universal relations is confirmed for those stars. Prospects of observations are also discussed., Comment: 13 pages, 10 figures

The observed timing data, magnetic tilt angle $\chi$, and age of young pulsars could be used to probe some important issues about neutron star (NS) physics, e.g., the NS internal magnetic field configuration, and the number of precession cycles $\xi$. \textbf{Both} quantities are critical in studying the continuous gravitational wave emission from pulsars, and the latter generally characterizes the mutual interactions between superfluid neutrons and other particles in the NS interior. The timing behavior of pulsars can be influenced by the dipole field evolution, which instead of decaying, may increase with time. An increase in the dipole field may result from the re-emergence of the initial dipole field $B_{\rm d,i}$ that was buried into the NS interior shortly after the birth of the NS. In this work, the field re-emergence scenario $\xi$ and the internal field configuration of several young pulsars, as well as their $B_{\rm d, i}$ are investigated by assuming typical accreted masses $\Delta M$. Moreover, since the Crab pulsar has an exactly known age and its tilt angle change rate can be inferred from observations, we can set stringent constraints on its $\xi$, $B_{\rm d,i}$, and $\Delta M$. Although for other young pulsars without exactly known ages and tilt angle change rates, these quantities cannot be accurately determined, we find that their $\xi$ are generally within $\sim10^4-10^6$, and some of them probably have magnetar-strength $B_{\rm d,i}$. Our work could be important for investigating the transient emissions associated with NSs, the origin of strong magnetic fields of NSs, pulsar population, continuous gravitational wave emission from pulsars, and accretion under extreme conditions in principle.

The initial spin periods of newborn magnetars are \textbf{strongly associated with the origin of their strong magnetic fields, both of which can affect the electromagnetic radiation and gravitational waves (GWs) emitted at their birth.} Combining the upper limit $E_{\rm SNR}\lesssim10^{51}$ erg on the explosion energies of \textbf{the supernova (SN) remnants around slowly-spinning magnetars} with a detailed investigation on the evolution of newborn magnetars, we set constraints on the initial spin periods of magnetars \textbf{born in weak SN explosions}. Depending on the conversion efficiency $\eta$ of the electromagnetic energy of \textbf{these} newborn magnetars into the kinetic energy of SN ejecta, the minimum initial spin periods of \textbf{these} newborn magnetars are $P_{\rm i, min}\simeq 5-6$ ms for an ideal efficiency $\eta=1$, $P_{\rm i, min}\simeq 3-4$ ms for a possible efficiency $\eta=0.4$, and $P_{\rm i, min}\simeq 1-2$ ms for a relatively low efficiency $\eta=0.1$. \textbf{Based on these constraints and adopting reasonable values for the physical parameters of the newborn magnetars, we find that their GW radiation at $\nu_{\rm e,1}=\nu$ may be undetectable by the Einstein Telescope (ET) since the maximum signal-to-noise ratio (${\rm S/N}$) is only 2.41 even the sources are located at a very close distance of 5 Mpc, where $\nu$ are the spin frequencies of the magnetars. At such a distance, the GWs emitted at $\nu_{\rm e,2}=2\nu$ from the newborn magnetars with dipole fields $B_{\rm d}=5\times10^{14}$ and $10^{15}$ G may be detectable by the ET because ${\rm S/N}$ are 10.01 and 19.85, respectively. However, if these newborn magnetars are located at $20$ Mpc away in the Virgo supercluster, no GWs could be detected by the ET due to low ${\rm S/N}$.

The observational data (e.g., the timing data and magnetic tilt angles $\chi$) of young pulsars can be used to probe some critical issues about the internal physics of neutron stars (NSs), for instance, the number of precession cycles $\xi$ and the internal magnetic field configuration (IMFC) of NSs. Evolution of the dipole magnetic field $B_{\rm d}$ of NSs may play an important role in determining the final results. In this work, a power-law form is adopted to describe the decay of $B_{\rm d}$. In such a scenario, the IMFC and $\xi$ of young pulsars with an ordinary $B_{\rm d}\sim10^{12}-\-10^{13}$ G and a steady braking index $n$ are investigated. Since the tilt angle change rates $\dot{\chi}$ of pulsars with $n<3$ can be theoretically predicted, a test on the power-law decay model can thus be made by comparing the theoretical values to that obtained from observations. However, such a comparison can only be made on the Crab pulsar currently, and the results show that the power-law decay model is inconsistent with the Crab's observations. We suggest that rather than decay, the Crab's $B_{\rm d}$ should increase with time at a rate $\sim12-14$ G/s. A definite conclusion on the validity of the power-law decay model for pulsars with ordinary $B_{\rm d}$ may be given if $\dot{\chi}$ of other pulsars could be measured.

Dissipation is usually considered a negative factor for observing quantum effects and for harnessing them for quantum technologies. Here we propose a scheme for speeding up the generation of quantum entanglement between two coupled qubits by introducing a strong dissipation channel to one of these qubits. The maximal entanglement is conditionally established by evenly distributing a single excitation between these two qubits. When the excitation is initially held by the dissipative qubit, the dissipation accelerates the excitation re-distribution process for the quantum state trajectory without quantum jumps. Our results show that the time needed to conditionally attain the maximal entanglement is monotonously decreased as the dissipative rate is increased. We further show that this scheme can be generalized to accelerate the production of the W state for the three-qubit system, where one NH qubit is symmetrically coupled to two Hermitian qubits., Comment: 2 figures

The Crab Pulsar has exhibited numerous glitches accompanied by persistent shifts in its spin-down rate. The explanation of the observed persistent shifts remain a challenge. We perform a detailed Bayesian analysis to compare four data-fitting models, ranging from a simple linear model to more complex power-law and logarithmic models, using a dataset of observed glitches and persistent shifts. Our results show the large observed events are difficult to explain by the usually assumed linear model due to starquakes. A particularly notable finding is that the logarithmic model provides the best fit to the observation data but the two power-law models show a close tie to it. Detail differences of these models may be further clarified by the understanding of internal physics of neutron stars., Comment: Submitted to ApJ

Graphene has garnered significant attention due to its unique properties. Among its many intriguing characteristics, the tuning effects induced by adsorbed atoms (adatoms) provide immense potential for the design of graphene-based electronic devices. This work explores the electronic migration in the adatom-graphene composite, using the extended dissipaton-equation-of-motion (DEOM) approach. As an exact dynamics theory for open quantum systems embedded in environments composed of non-interacting electrons, the extended DEOM is capable of handling both linear and quadratic environmental couplings (a certain non-Gaussian effect) which account for the interactions between the adatom and the graphene substrate. We demonstrate and analyze the adatom-graphene correlated properties and the tuning effects by simulating the adatom spectral functions with varied Coulomb repulsion strengths. This work offers not only advanced theoretical methods but also new insights into the theoretical investigation of complex functional materials such as graphene-based electronic devices., Comment: 8 pages, 5 figures

In many RFID-enabled applications, objects are classified into different categories, and the information associated with each object's category (called category information) is written into the attached tag, allowing the reader to access it later. The category information sampling in such RFID systems, which is to randomly choose (sample) a few tags from each category and collect their category information, is fundamental for providing real-time monitoring and analysis in RFID. However, to the best of our knowledge, two technical challenges, i.e., how to guarantee a minimized execution time and reduce collection failure caused by missing tags, remain unsolved for this problem. In this paper, we address these two limitations by considering how to use the shortest possible time to sample a different number of random tags from each category and collect their category information sequentially in small batches. In particular, we first obtain a lower bound on the execution time of any protocol that can solve this problem. Then, we present a near-OPTimal Category information sampling protocol (OPT-C) that solves the problem with an execution time close to the lower bound. Finally, extensive simulation results demonstrate the superiority of OPT-C over existing protocols, while real-world experiments validate the practicality of OPT-C., Comment: 37 pages, 11 figures

Reconfigurable intelligent surface (RIS), composed of nearly passive elements, is regarded as one of the potential paradigms to support multi-gigabit data in real-time. However, in traditional CSI (channel state information) driven frame, the training overhead of channel estimation greatly increases as the number of RIS elements increases to intelligently manipulate the reflected signals. To conveniently use the reflected signal without complex CSI feedback, in this paper we propose a position-aided phase configuration scheme based on the property of Fresnel zone. In particular, we design the impedance based discrete RIS elements with joint absorption mode and reflection mode considering the fabrication complexities, which integrated the property of the Fresnel zone to resist the impact of position error. Then, with joint absorption and 1-bit reflection mode elements, we develop the two-step position-aided ON/OFF states judgement (TPOSJ) scheme and the frame structure to control the ON/OFF state of RIS, followed by analyzing the impacts of mobility and position error on our proposed scheme. Also, we derive the Helmholtz-Kirchhoff integral theorem based power flow. Simulations show that the proposed scheme can manipulate the ON/OFF state intelligently without complex CSI, thus verifying the practical application of our proposed scheme., Comment: 15 pages, accepted by IEEE TCOM (early access)

Owing to its outstanding parallel computing capabilities, quantum computing (QC) has been a subject of continuous attention. With the gradual maturation of QC platforms, it has increasingly played a significant role in various fields such as transportation, pharmaceuticals, and industrial manufacturing,achieving unprecedented milestones. In modern society, wireless communication stands as an indispensable infrastructure, with its essence lying in optimization. Although artificial intelligence (AI) algorithms such as Reinforcement Learning (RL) and mathematical optimization have greatly enhanced the performance of wireless communication, the rapid attainment of optimal solutions for wireless communication problems remains an unresolved challenge. QC, however, presents a new alternative. This paper aims to elucidate the fundamentals of QC and explore its applications in wireless communications and networking. First, we will provide a tutorial on QC, covering its basics, characteristics, and popular QC algorithms. Next, we will introduce the applications of QC in communication and networking, followed by its applications in miscellaneous areas such as security and privacy, localization and tracking, and video streaming. Finally,we will discuss remaining open issues before concluding.

Future wireless communications are promising to support ubiquitous connections and high data rates with cost-effective devices. Benefiting from the energy-efficient elements with low cost, reconfigurable intelligent surface (RIS) emerges as a potential solution to fulfill such demands, which has the capability to flexibly manipulate the wireless signals with a tunable phase. Recently, as the operational frequency ascends to the sub-terahertz (THz) bands or higher bands for wireless communications in six-generation (6G), it becomes imperative to consider the near-field propagation in RIS-assisted communications. The challenging acquisition of channel parameters is an inherent issue for near-field RIS-assisted communications, where the complex design is essential to acquire the informative near-field channel embedded with both the angle and distance information. Hence, in this paper we systematically investigate the potential of exploiting location information for near-field RIS-assisted communications. Firstly, we present the progresses in the near-field RIS-assisted communications, which are compatible with existing wireless communications and show the potential to achieve the fine-grained localization accuracy to support location-driven scheme. Then, the Fresnel zone based model is introduced, with which the location-driven beamforming scheme and corresponding frame structure are developed. Also, we elaborate on four unique advantages for leveraging location information in RIS-assisted communications, followed by numerical simulations. Finally, several key challenges and corresponding potential solutions are pointed out., Comment: 7 pages, 6 figures, accepted by IEEE Communication Magazine

Two-color laser induced plasma filamentation in air could serve as tabletop sources of broadband terahertz (THz) pulses. Ubiquitous air in the earth facilitates its widespread utilities, particularly, in wireless communication and remote sensing, exploiting the unique advantage of the air-based scheme that the THz sources can be delivered over standoff distances via the pump laser propagation in air. However, the THz emission pattern inevitably has a conical angular profile with a dip in the propagation axis, therefore, THz energy concentration, propagation directionality, and accuracy of signal demodulation are significantly impaired, greatly limiting its direct applications. Here, we successfully eliminate the unfavorable conical profile by experiments and meanwhile enhance THz directionality and intensity by 17 folds by use of abruptly-autofocusing laser beams. Our theory and simulations show that these observations are attributed to efficient suppression of the dephasing effect appearing in previous investigations with ordinary Gaussian laser beams. This scheme is easily accessible since the abruptly-autofocusing beam can be achieved by imposing a spatial light modulator on the input Gaussian beam. This study solves a long-standing problem in the two-color laser scheme and clarifies the underlying physics of the conical angular profile formation., Comment: 22 pages, 4 figures, 1 table

Simulating the dynamics of open quantum systems coupled to non-Markovian environments remains an outstanding challenge due to exponentially scaling computational costs. We present an artificial intelligence strategy to overcome this obstacle by integrating the neural quantum states approach into the dissipaton-embedded quantum master equation in second quantization (DQME-SQ). Our approach utilizes restricted Boltzmann machines (RBMs) to compactly represent the reduced density tensor, explicitly encoding the combined effects of system-environment correlations and nonMarkovian memory. Applied to model systems exhibiting prominent effects of system-environment correlation and non-Markovian memory, our approach achieves comparable accuracy to conventional hierarchical equations of motion, while requiring significantly fewer dynamical variables. The novel RBM-based DQME-SQ approach paves the way for investigating non-Markovian open quantum dynamics in previously intractable regimes, with implications spanning various frontiers of modern science., Comment: 7 pages, 5 figures

Recently, signatures of superconductivity with critical temperature from 20 to 30 K have been reported in pressured trilayer nickelate La$_4$Ni$_3$O$_{10}$ through a pressure-induced structure transition. Here we explore the evolution of electronic structures and electronic correlations in different phases of La$_4$Ni$_3$O$_{10}$ under corresponding pressure regions, by using density functional theory (DFT) combined with dynamical mean-field theory (DMFT). Similar to bilayer superconductor La$_3$Ni$_2$O$_{7}$, the electronic bands in superconducting La$_4$Ni$_3$O$_{10}$ are dominated by Ni-3$d_{x^2-y^2}$ and 3$d_{z^2}$ orbits near the Fermi level, in contrast, the inner Ni-O plane in La$_4$Ni$_3$O$_{10}$ generates a doublet hole-pocket Fermi surfaces around the Brillouin-zone corner, meanwhile one branch of the Ni-$3d_{z^2}$ bands is pushed very close above the Fermi level, which can induce an electron pocket through small electron doping. The DFT+DMFT simulations suggest that the electronic correlations only give minor modification to the Fermi surfaces, meanwhile the Ni-$3d_{z^2}$ and 3$d_{x^2-y^2}$ states on outer Ni-O layers have considerable greater mass enhancements than on the inner layer. The sensitiveness of electronic structure under doping and unique layer dependence of correlation suggest a distinct superconducting mechanism with respect to bilayer La$_3$Ni$_2$O$_{7}$. Based on the DFT and DFT+DMFT simulations, we eventually derive a trilayer effective tight-binding model, which can produce rather precise electronic bands and Fermi surfaces, hence can serve as an appropriate model to further study the superconducting mechanism and paring symmetry in trilayer La$_4$Ni$_3$O$_{10}$., Comment: 8 pages, 10 figures

Astigmatic unitary transformations allow for the adiabatic connections of all feasible states of paraxial Gaussian beams on the same modal sphere, i.e., Hermite-Laguerre-Gaussian (HLG) modes. Here, we present a comprehensive investigation into the unitary modal evolution of complex structured Gaussian beams, comprised by HLG modes from disparate modal spheres, via astigmatic transformation. The non-synchronized higher-order geometric phases in cyclic transformations originates a Talbot-effect-like modal evolution in the superposition state of these HLG modes, resulting in pattern variations and revivals in transformations with specific geodesic loops. Using Ince-Gaussian modes as an illustrative example, we systematically analyze and experimentally corroborate the beamforming mechanism behind the pattern evolution. Our results outline a generic modal conversion theory of structured Gaussian beams via astigmatic unitary transformation, offering a new approach for shaping spatial modal structure. These findings may inspire a wide variety of applications based on structured light., Comment: 7 pages, 5 figures

The integration of scanning tunneling microscopy (STM) and electron spin resonance (ESR) spectroscopy has emerged as a powerful and innovative tool for discerning spin excitations and spin-spin interactions within atoms and molecules adsorbed on surfaces. However, the origin of the STM-ESR signal and the underlying mechanisms that govern the essential features of the measured spectra have remained elusive, thereby significantly impeding the future development of the STM-ESR approach. Here, we construct a model to carry out precise numerical simulations of STM-ESR spectra for a single hydrogenated Ti adatom and a hydrogenated Ti dimer, achieving excellent agreement with experimental observations. We further develop an analytic theory that elucidates the fundamental origin of the signal as well as the essential features in the measured spectra. These new theoretical developments establish a solid foundation for the on-demand detection and manipulation of atomic-scale spin states, with promising implications for cutting-edge applications in spin sensing, quantum information, and quantum computing., Comment: 6 pages, 4 figures

The CMS and ATLAS have recently reported their results searching for light Higgs boson with mass around 95 GeV, based on the full Run 2 data set. In the framework of the CP-violating (CPV) $\mu\nu$SSM, we discuss a $\sim$ 2.9$\sigma$ (local) excess at 95 GeV in the light Higgs boson search in the diphoton decay mode as reported by ATLAS and CMS, together with a $\sim$ 2$\sigma$ excess (local) in the $b\bar{b}$ final state at LEP in the same mass range. By introducing CPV phases as well as by mixing CP-even Higgs and CP-odd Higgs, a lighter Higgs boson in the $\mu\nu$SSM can be produced, which can account for the "di-photon excess"., Comment: 29 pages, 7 figures, accepted for publication in Physical Review D

Non-Markovianity, the intricate dependence of an open quantum system on its temporal evolution history, holds tremendous implications across various scientific disciplines. However, accurately characterizing the complex non-Markovian effects has posed a formidable challenge for numerical simulations. While quantum computing technologies show promise, a universal theory enabling practical quantum algorithm implementation has been elusive. We address this gap by introducing the dissipaton-embedded quantum master equation in second quantization (DQME-SQ). This exact and compact theory offers two key advantages: representability by quantum circuits and universal applicability to any Gaussian environment. We demonstrate these capabilities through digital quantum simulations of non-Markovian dissipative dynamics in both bosonic and fermionic environments. The DQME-SQ framework opens a new horizon for the efficient exploration of complex open quantum systems by leveraging the rapidly advancing quantum computing technologies., Comment: 14 pages, 5 figures

Many-body open quantum systems (OQS) have a profound impact on various subdisciplines of physics, chemistry, and biology. Thus, the development of a computer program capable of accurately, efficiently, and versatilely simulating many-body OQS is highly desirable. In recent years, we have focused on the advancement of numerical algorithms based on the fermionic hierarchical equations of motion (HEOM) theory. Being in-principle exact, this approach allows for the precise characterization of many-body correlations, non-Markovian memory, and non-equilibrium thermodynamic conditions. These efforts now lead to the establishment of a new computer program, HEOM for QUantum Impurity with a Correlated Kernel, version 2 (HEOM-QUICK2), which, to the best of our knowledge, is currently the only general-purpose simulator for fermionic many-body OQS. Compared with version 1, the HEOM-QUICK2 program features more efficient solvers for stationary states, more accurate treatment of non-Markovian memory, and improved numerical stability for long-time dissipative dynamics. Integrated with quantum chemistry software, HEOM-QUICK2 has become a valuable theoretical tool for the precise simulation of realistic many-body OQS, particularly the single atomic or molecular junctions. Furthermore, the unprecedented precision achieved by HEOM-QUICK2 enables accurate simulation of low-energy spin excitations and coherent spin relaxation. The unique usefulness of HEOM-QUICK2 is demonstrated through several examples of strongly correlated quantum impurity systems under non-equilibrium conditions. Thus, the new HEOM-QUICK2 program offers a powerful and comprehensive tool for studying many-body OQS with exotic quantum phenomena and exploring applications in various disciplines., Comment: 67 pages; 13 figures

Despite tremendous advances in AI, it remains a significant challenge to develop interactive task guidance systems that can offer situated, personalized guidance and assist humans in various tasks. These systems need to have a sophisticated understanding of the user as well as the environment, and make timely accurate decisions on when and what to say. To address this issue, we created a new multimodal benchmark dataset, Watch, Talk and Guide (WTaG) based on natural interaction between a human user and a human instructor. We further proposed two tasks: User and Environment Understanding, and Instructor Decision Making. We leveraged several foundation models to study to what extent these models can be quickly adapted to perceptually enabled task guidance. Our quantitative, qualitative, and human evaluation results show that these models can demonstrate fair performances in some cases with no task-specific training, but a fast and reliable adaptation remains a significant challenge. Our benchmark and baselines will provide a stepping stone for future work on situated task guidance., Comment: Accepted to EMNLP 2023 Findings

Black widows (BWs) are millisecond pulsars ablating their companion stars. The out-flowing material from the companion can block the radio emission of the pulsar, resulting in eclipses. In this paper, we construct a model for the radio eclipse by calculating the geometry of the bow shock between the winds of the pulsar and companion, where the shock shapes the eclipsing medium but had not been described in detail in previous works. The model is further used to explain the variations of the flux density and dispersion measure (DM) of three BW pulsars (i.e., PSR B1957$+$20, J2055$+$3829, and J2051$-$0827) detected by the Five-hundred-meter Aperture Spherical radio Telescope (FAST). Consequently, we constrained the parameters of the three BW systems such as the inclination angles and true anomalies of the observer as well as the mass-loss rates and wind velocity of the companion stars. With the help of these constraints, it is expected that magnetic fields of companion stars and even masses of pulsars could further be determined as some extra observation can be achieved in the future., Comment: 12 pages, 6 figures, accepted for publication in RAA

In many Ce-based superconductors, superconducting (SC) phases emerge or can be tuned in proximity to the antiferromagnetic (AF) quantum critical point (QCP), but so far the explicit phase evolution near the QCP lack theoretical understanding. Here, by combing the density functional theory plus dynamical mean-field theory (DFT+DMFT) with effective-model calculations, we provide a theoretical description for Ce-based superconductors under compression. DFT+DMFT calculations for the normal states reveal that the Kondo hybridizations are significantly enhanced under compression, while the initially localized $f$ electrons become fully itinerant via localized-itinerant crossover. We then construct an effective model and show that with the extracted Kondo coupling and RKKY exchange strengths from first-principle calculations, the ground-state phases of these materials can be properly predicted. We also show that the coexistence of magnetic correlation and Kondo hybridization can drive AF+SC coexisting state in narrow compression region. Under compression, competition between Kondo and RKKY interactions can drive successive transitions, from AF phase to AF+SC coexisting phase, then to paramagnetic SC phase via an AF transition which generates the QCP, and finally to normal Kondo paramagnetic (KP) phase through an SC-KP transition induced by the localized-itinerant crossover. Our study gives proper explanation to the pressure-induced QCP and SC-KP transition, and to the phase evolution in pressured Ce-based superconductors, and can help to understand the SC states around the ferromagnetic quantum transition points in uranium-based superconductors., Comment: 16 pages, 14 figures

Microservice architecture has sprung up over recent years for managing enterprise applications, due to its ability to independently deploy and scale services. Despite its benefits, ensuring the reliability and safety of a microservice system remains highly challenging. Existing anomaly detection algorithms based on a single data modality (i.e., metrics, logs, or traces) fail to fully account for the complex correlations and interactions between different modalities, leading to false negatives and false alarms, whereas incorporating more data modalities can offer opportunities for further performance gain. As a fresh attempt, we propose in this paper a semi-supervised graph-based anomaly detection method, MSTGAD, which seamlessly integrates all available data modalities via attentive multi-modal learning. First, we extract and normalize features from the three modalities, and further integrate them using a graph, namely MST (microservice system twin) graph, where each node represents a service instance and the edge indicates the scheduling relationship between different service instances. The MST graph provides a virtual representation of the status and scheduling relationships among service instances of a real-world microservice system. Second, we construct a transformer-based neural network with both spatial and temporal attention mechanisms to model the inter-correlations between different modalities and temporal dependencies between the data points. This enables us to detect anomalies automatically and accurately in real-time. The source code of MSTGAD is publicly available at https://github.com/alipay/microservice_system_twin_graph_based_anomaly_detection., Comment: This paper have been accepted by ASE 2023

The rapid adoption of AI and convenience offered by cloud services have resulted in the growing demands for GPUs in the cloud. Generally, GPUs are physically attached to host servers as PCIe devices. However, the fixed assembly combination of host servers and GPUs is extremely inefficient in resource utilization, upgrade, and maintenance. Due to these issues, the GPU disaggregation technique has been proposed to decouple GPUs from host servers. It aggregates GPUs into a pool, and allocates GPU node(s) according to user demands. However, existing GPU disaggregation systems have flaws in software-hardware compatibility, disaggregation scope, and capacity. In this paper, we present a new implementation of datacenter-scale GPU disaggregation, named DxPU. DxPU efficiently solves the above problems and can flexibly allocate as many GPU node(s) as users demand. In order to understand the performance overhead incurred by DxPU, we build up a performance model for AI specific workloads. With the guidance of modeling results, we develop a prototype system, which has been deployed into the datacenter of a leading cloud provider for a test run. We also conduct detailed experiments to evaluate the performance overhead caused by our system. The results show that the overhead of DxPU is less than 10%, compared with native GPU servers, in most of user scenarios., Comment: 23 pages, 6 figures, published in ACM Transactions on Architecture and Code Optimization